* src/Makefile.am (cu_install_program): New variable, set to either ./ginstall or @INSTALL_PROGRAM@ (for cross compilation). (INSTALL_PROGRAM): Override AC_SUBST unconditionally, to avoid warning, and wrong initialization order. ---
D'oh. If and AC_SUBSTed value is not overridden unconditionally (here: INSTALL_PROGRAM), then the AC_SUBSTed value seems to always win. At least that is what my tests say. I don't fully understand the reasoning behind automake/tests/cond2[34].test <http://lists.gnu.org/archive/html/automake/2002-06/msg00157.html> <http://lists.gnu.org/archive/html/automake-patches/2002-06/msg00066.html> yet, but now that I understand the semantics, they are easily enough overridden. So this patch also has the advantage to fix the warning that automake gave. Cheers, Ralf src/Makefile.am | 7 +++++-- 1 files changed, 5 insertions(+), 2 deletions(-) diff --git a/src/Makefile.am b/src/Makefile.am index 80313ae..3625878 100644 --- a/src/Makefile.am +++ b/src/Makefile.am @@ -466,6 +466,9 @@ sc_tight_scope: $(all_programs) exit 1; } || : # Use the just-built ./ginstall, when not cross-compiling. -if !CROSS_COMPILING -INSTALL_PROGRAM = ./ginstall +if CROSS_COMPILING +cu_install_program = @INSTALL_PROGRAM@ +else +cu_install_program = ./ginstall endif +INSTALL_PROGRAM = $(cu_install_program) -- 1.6.0.1.309.g48068 _______________________________________________ Bug-coreutils mailing list [email protected] http://lists.gnu.org/mailman/listinfo/bug-coreutils
